Over the last five years, drug design and development has evolved substantially with technological innovations. Electronic health records (EHRs) have played an important role in utilising health data on a digital platform in both preclinical and clinical studies. Biobanks have also proven to be useful in enabling extensive genomic studies. These are a few examples of recent progress in the design and development of pharmacological drugs.

For daily articles on the latest pharma trends and innovations, as well as interviews with leading experts and in-depth industry White Papers, subscribe to PharmaFeatures.com.

A biobank is generally defined as a “collection of human biological samples and associated information organized in a systematic way for research purposes”. Biobanks have been implemented across many areas including drug development and clinical trial monitoring for predicting drug effects. Consent is mandatory for participation in the UK Biobank, based on an explanation and understanding of the process. The following are a few examples of the what a patient will be consenting to, as described in a document by the UK Biobank:

• The purpose of the UK Biobank, the fact that it is a long term research resource (not a healthcare programme), and any risks and benefits of taking part

• The kinds of information and samples that will be collected at enrolment, which may include  data that some participants consider especially sensitive

• The expectation that commercial entities will apply to use UK Biobank

Biobanks have a wide range of uses, including the discovery of important genotype-phenotype relationships. Phenome-wide associations (PheWas) studies utilise the large-scale biobanks, beginning with a genotype and analysing vast numbers of phenotypes for disease-gene associations. 

According to a 2019 review, the PheWas approach has been used to support the target validation for common disease indications, detecting possible pleiotropy. Pleiotropy refers to the association of multiple phenotypes with a single variant. Identifying disease-gene associations is critical in drug development, as it helps to decipher the mechanism of diseases from genetic variations to the manifestation of the phenotype. In addition to PheWas studies, modelling variants in an allelic series as a dose-response provides target validation. 

Up until several years ago, the vast amount of data in biobanks presented an issue for technology that struggled to process huge genetic datasets. Now, next-generation sequencing is used to rapidly sequence the extensive genetic data within the UK biobanks. 

Challenges

Despite evidence supporting their utility in drug design and development, there remain a number of concerns regarding the complicated ethical considerations of how the tissue samples are used. In order to protect the identity of donors, samples collected are often ‘de-identified’. This implies that biobanking is low risk and that de-identifying materials provides adequate protection to donors. 

However genomic investigation like phenome-wide association studies indirectly identify donors. Other forms of data collection from smartphones and wearable devices, for example, enable the sharing and analysis of data for healthcare and commercial applications. Of course donors submit an extensive consent form explaining how their sample may be used, but if material is collected without disclosure as stated above, this presents an ethical risk. 

Electronic health records

EHRs are real-time digital databases containing everything from patient medical history, clinical visits, medication and general healthcare. EHRs have proven to be useful across scientific and clinical research, whereby EHR-linked population-based biobanks have provided “significant opportunities for translational and implementation research that drive personalized medicine”.

Firstly, EHRs have been used for drug usage re-evaluation. The systematic screening of ‘off-label’ drugs can provide insights into overlaps between disease pathology, representing an innovative approach for generating new drug repositioning hypotheses. Off-label refers to the unapproved use of a drug, i.e.  a drug is used for a disease or medical condition that it is not approved to treat, such as when a chemotherapy is approved to treat one type of cancer, but healthcare providers use it to treat a different type of cancer. 

There are a number of reasons why off-label drugs are used. Often they are prescribed by doctors for patients whose disease has either seen no response to all available treatment options, or there is no approved treatment yet. 

There is also an increasing appreciation of the benefits of in-text mining EHR information for drug discovery. Text mining enables the computational analysis of text-locked data. This involves identification of specific text such as diseases, genes, or related terminology. The two core functions of text mining are the extraction of information and generation of hypotheses from extracted information about relationships. 

In precision oncology, in-text mining of EHRs has been implemented in preclinical and clinical applications. It involves “(1) characterization of the “driver” mutations in a given patient’s tumor; and (2) identification of the drugs that will best counteract the effects of those driver mutations”.  

Challenges 

As with the use of biobanks, one of the obvious challenges when using EHRs is the ethical concerns surrounding consent and confidentiality. Research organisations which require access to such records request the consent of the healthcare organisations storing them.

Firstly, even with legitimate use for access to EHRs, there are patient concerns about access to information they regard as confidential. For example, a patient may not want their ophthalmologist knowing about their mental health history. Although patients are asked to consent for comprehensive access, “the degree to which they fully understand the potential consequences of this consent, including the risk of security breaches and other implications for privacy, remains unclear.” 

Psychiatric history is a particular area for concern as some records may contain detailed, traumatic experiences associated with a mental health problem that a patient may wish to keep confidential from clinicians who do not require this information. One potential solution suggested is to implement a set of ‘safeguards’. This involves ‘soft barriers’ designed to remind providers to engage in ethical and appropriate use of the medical record. In addition to passwords or key cards, “break the glass” reminders that ask physicians to affirm their right to access sensitive material in the EHR before they do so. 

Secondly, as a web-based platform, EHRs are subject to security breaches from outside the system: whether this be the misplacement of a physician’s laptop or external hacking from criminal organisations. 

The safeguards mentioned previously have also been utilised for data security. Contingency plans and frequent auditing are examples of administrative safeguarding, one of three safeguarding themes. The second theme is physical safeguards which focus on compliant security procedures. This focuses on protecting the health information physically so that software or hardware is not accessed by unauthorised persons or those who could misuse them. The final category of security protection is technical safeguards, facilitated by firewalls, encryption, virus checking and measures used in authenticating information. 

The constant evolution of technology and data security will further optimise the use of biobanks and EHRs for drug development. With evidence supporting their role in the development of therapeutic treatment, it is likely future investigation into these databases will reveal more ways they can be used in drug development.  

To discuss these topics further with sector experts, and to ensure you remain up-to-date on the latest in clinical development, sign up for Proventa International’s Bioinformatics Strategy Meeting, set for 1 July 2021.

Charlotte Di Salvo, Junior Medical Writer
Proventa International